def test_set_children_raises(self):
test = self._test_set_children_raises
test(Word("foo"), [Word("foo")])
test(Phrase('"foo"'), [Word("foo")])
test(Regex("/foo/"), [Word("foo")])
test(SearchField("foo", Word("bar")), [])
test(SearchField("foo", Word("bar")), [Word("bar"), Word("baz")])
test(Group(Word("foo")), [])
test(Group(Word("foo")), [Word("foo"), Word("bar")])
test(FieldGroup(Word("foo")), [])
test(FieldGroup(Word("foo")), [Word("foo"), Word("bar")])
test(Range(Word("20"), Word("30")), [])
test(Range(Word("20"), Word("30")),
[Word("20"), Word("30"), Word("40")])
test(Proximity(Word("foo")), [])
test(Proximity(Word("foo")), [Word("foo"), Word("bar")])
test(Fuzzy(Word("foo")), [])
test(Fuzzy(Word("foo")), [Word("foo"), Word("bar")])
test(Boost(Word("foo"), force=1), [])
test(Boost(Word("foo"), force=1), [Word("foo"), Word("bar")])
test(Plus(Word("foo")), [])
test(Plus(Word("foo")), [Word("foo"), Word("bar")])
test(Not(Word("foo")), [])
test(Not(Word("foo")), [Word("foo"), Word("bar")])
test(Prohibit(Word("foo")), [])
test(Prohibit(Word("foo")), [Word("foo"), Word("bar")])
def test_approx(self):
tree = (UnknownOperation(
Proximity(Phrase('"foo bar"'), 3, tail=" "),
Proximity(Phrase('"foo baz"'), None, tail=" "),
Fuzzy(Word('baz'), Decimal("0.3"), tail=" "),
Fuzzy(Word('fou'), None)))
parsed = parser.parse('"foo bar"~3 "foo baz"~ baz~0.3 fou~')
self.assertEqual(str(parsed), str(tree))
self.assertEqual(parsed, tree)
def test_fuzzy(self):
orig = Fuzzy(Word("bar"), degree=.3, pos=3, head="\n", tail="\t")
copy = orig.clone_item()
self.assert_equal_tail_head_pos(orig, copy)
self.assertEqual(orig.degree, copy.degree)
self.assertEqual(copy.term, NONE_ITEM)
self.assertNotEqual(orig, copy)
copy.children = [Word("bar")]
self.assertEqual(orig, copy)
def test_equality_fuzzy(self):
f1 = Fuzzy(term=Word('foo'), degree=5)
f2 = Fuzzy(term=Word('bar'), degree=5)
f3 = Fuzzy(term=Word('foo'), degree=5)
f4 = Fuzzy(term=Word('foo'), degree=.5)
f5 = Fuzzy(term=Word('foo'), degree=None)
self.assertNotEqual(f1, f2)
self.assertEqual(f1, f3)
self.assertNotEqual(f1, f4)
self.assertEqual(f4, f5)
def test_fuzzy(self):
item = Fuzzy(Word("foo"), degree=None)
self.assertEqual(str(item), "foo~")
self.assertEqual(repr(item), "Fuzzy(Word('foo'), 0.5)")
self.assertEqual(item.degree, Decimal(".5").normalize())
item = Fuzzy(Word("foo"), degree=".5")
self.assertEqual(str(item), "foo~0.5")
item = Fuzzy(Word("foo"), degree=str(1 / 3))
self.assertEqual(str(item), "foo~0.3333333333333333")
# head tail
item = Fuzzy(Word("foo", head="\t", tail="\n"), head="\r", tail=" ")
self.assertEqual(str(item), "\tfoo\n~")
self.assertEqual(item.__str__(head_tail=True), "\r\tfoo\n~ ")
def test_auto_name_nested(self):
tree = AndOperation(
OrOperation(
SearchField("bar", Word("test")),
AndOperation(
Proximity(Phrase('"test"'), 2),
SearchField("baz", Word("test")),
),
),
Group(UnknownOperation(
Fuzzy(Word("test")),
Phrase('"test"'),
), ),
)
names = auto_name(tree)
self.assertEqual(sorted(names.keys()), list("abcdefgh"))
# and
and1 = tree
self.assertEqual(get_name(and1), None)
# - or
or1 = and1.children[0]
self.assertEqual(get_name(or1), "a")
self.assertEqual(names["a"], (0, ))
# -- search field
sfield1 = or1.children[0]
self.assertEqual(get_name(sfield1), "c")
self.assertEqual(names["c"], (0, 0))
self.assertEqual(get_name(sfield1.expr), None)
# -- and
and2 = or1.children[1]
self.assertEqual(get_name(and2), "d")
self.assertEqual(names["d"], (0, 1))
# --- proximity phrase
self.assertEqual(get_name(and2.children[0]), "e")
self.assertEqual(names["e"], (0, 1, 0))
self.assertEqual(get_name(and2.children[0].term), None)
# --- search field
sfield2 = and2.children[1]
self.assertEqual(get_name(sfield2), "f")
self.assertEqual(names["f"], (0, 1, 1))
self.assertEqual(get_name(sfield2.expr), None)
# - group
group1 = and1.children[1]
self.assertEqual(get_name(group1), "b")
self.assertEqual(names["b"], (1, ))
# -- unknown op
unknownop1 = group1.children[0]
self.assertEqual(get_name(unknownop1), None)
# --- fuzzy word
self.assertEqual(get_name(unknownop1.children[0]), "g")
self.assertEqual(names["g"], (1, 0, 0))
self.assertEqual(get_name(unknownop1.children[0].term), None)
# --- phrase
self.assertEqual(get_name(unknownop1.children[1]), "h")
self.assertEqual(names["h"], (1, 0, 1))
def test_visit_complex(self):
tree = AndOperation(
Group(
OrOperation(Word("foo"), Word("bar"),
Boost(Fuzzy(Word("baz")), force=2))),
Proximity(Phrase('"spam ham"')),
SearchField("fizz", Regex("/fuzz/")),
)
transformed = self.transform(tree)
expected = AndOperation(
Group(
OrOperation(
Word("foo@0-0-0"),
Word("bar@0-0-1"),
Boost(Fuzzy(Word("baz@0-0-2-0-0")), force=2),
)),
Proximity(Phrase('"spam ham@1-0"')),
SearchField("fizz", Regex("/fuzz@2-0/")),
)
self.assertEqual(transformed, expected)
def test_fuzzy(self):
tree = parser.parse("\rfoo\t~2\n")
self.assertEqual(tree, Fuzzy(Word("foo"), 2))
self.assertEqual(tree.head, "")
self.assertEqual(tree.tail, "\n")
self.assertEqual(tree.pos, 0)
self.assertEqual(tree.size, 7)
foo, = tree.children
self.assertEqual(foo.head, "\r")
self.assertEqual(foo.tail, "\t")
self.assertEqual(foo.pos, 1)
self.assertEqual(foo.size, 3)
self.assertEqual(str(tree), "\rfoo\t~2")
self.assertEqual(tree.__str__(head_tail=True), "\rfoo\t~2\n")
def test_element_from_path(self):
tree = AndOperation(
OrOperation(
SearchField("bar", Word("test")),
Group(
AndOperation(
Proximity(Phrase('"test"'), 2),
SearchField("baz", Word("test")),
Fuzzy(Word("test")),
Phrase('"test"'),
), ),
), )
names = {
"a": (),
"b": (0, 1),
"c": (0, 1, 0, 2),
"d": (0, 1, 0, 2, 0),
"e": (0, 1, 0, 3)
}
self.assertEqual(element_from_path(tree, ()), tree)
self.assertEqual(element_from_name(tree, "a", names), tree)
self.assertEqual(element_from_path(tree, (0, 1)),
tree.children[0].children[1])
self.assertEqual(element_from_name(tree, "b", names),
tree.children[0].children[1])
self.assertEqual(element_from_path(tree, (0, 1, 0, 2)),
Fuzzy(Word("test")))
self.assertEqual(element_from_name(tree, "c", names),
Fuzzy(Word("test")))
self.assertEqual(element_from_path(tree, (0, 1, 0, 2, 0)),
Word("test"))
self.assertEqual(element_from_name(tree, "d", names), Word("test"))
self.assertEqual(element_from_path(tree, (0, 1, 0, 3)),
Phrase('"test"'))
self.assertEqual(element_from_name(tree, "e", names), Phrase('"test"'))
with self.assertRaises(IndexError):
element_from_path(tree, (1, ))
def test_check_ok(self):
query = (AndOperation(
SearchField(
"f",
FieldGroup(
AndOperation(
Boost(Proximity(Phrase('"foo bar"'), 4), "4.2"),
Prohibit(Range("100", "200"))))),
Group(OrOperation(Fuzzy(Word("baz"), ".8"), Plus(Word("fizz"))))))
check = LuceneCheck()
self.assertTrue(check(query))
self.assertEqual(check.errors(query), [])
check = LuceneCheck(zeal=1)
self.assertTrue(check(query))
self.assertEqual(check.errors(query), [])
def test_named_queries_fuzzy(self):
tree = SearchField("text", Fuzzy(Word('bar')))
set_name(tree.children[0], "a")
result = self.transformer(tree)
self.assertEqual(
result,
{
"fuzzy": {
"text": {
"value": "bar",
"_name": "a",
'fuzziness': 0.5
}
}
},
)
def test_visit_complex(self):
tree = AndOperation(
Group(
OrOperation(Word("foo"), Word("bar"),
Boost(Fuzzy(Word("baz")), force=2))),
Proximity(Phrase('"spam ham"')),
SearchField("fizz", Regex("/fuzz/")),
)
paths = self.visit(tree)
self.assertEqual(sorted(paths), [
((0, 0, 0), "foo"),
((0, 0, 1), "bar"),
((0, 0, 2, 0, 0), "baz"),
((1, 0), '"spam ham"'),
((2, 0), '/fuzz/'),
])
def test_no_propagation(self):
for tree in [
Range(Word("a"), Word("b")),
Fuzzy(Word("foo")),
Proximity('"bar baz"', 2)
]:
with self.subTest("%r" % type(tree)):
paths_ok, paths_ko = self.propagate_matching(tree, set(), {()})
# no down propagation
self.assertEqual(paths_ok, set())
self.assertEqual(paths_ko, {()})
paths_ok, paths_ko = self.propagate_matching(tree, {()}, set())
self.assertEqual(paths_ok, {()})
self.assertEqual(paths_ko, set())
def test_auto_name_nested(self):
tree = AndOperation(
OrOperation(
SearchField("bar", Word("test")),
AndOperation(
Proximity(Phrase('"test"'), 2),
SearchField("baz", Word("test")),
),
),
Group(
UnknownOperation(
Fuzzy(Word("test")),
Phrase('"test"'),
),
),
)
auto_name(tree)
# and
and1 = tree
self.assertEqual(get_name(and1), "0")
# - or
or1 = and1.children[0]
self.assertEqual(get_name(or1), "0_0")
# --- search field word
sfield1 = or1.children[0]
self.assertFalse(get_name(sfield1))
self.assertEqual(get_name(sfield1.expr), "0_0_0")
# --- and
and2 = or1.children[1]
self.assertEqual(get_name(and2), "0_0_1")
# ----- proximity phrase
self.assertEqual(get_name(and2.children[0].term), "0_0_1_0")
# ----- search field word
sfield2 = and2.children[1]
self.assertFalse(get_name(sfield2))
self.assertEqual(get_name(sfield2.expr), "0_0_1_1")
# - group
group1 = and1.children[1]
self.assertEqual(get_name(group1), None)
# --- unknown op
unknownop1 = group1.children[0]
self.assertEqual(get_name(unknownop1), "0_1")
# ----- fuzzy word
self.assertEqual(get_name(unknownop1.children[0].term), "0_1_0")
# ----- phrase
self.assertEqual(get_name(unknownop1.children[1]), "0_1_1")
def test_complex(self):
tree = Group(
OrOperation(
SearchField(
"foo",
FieldGroup(
UnknownOperation(Word("bar"),
Range(Word("baz"), Word("spam")))),
),
Not(Proximity(Phrase('"ham ham"'), 2)),
Plus(Fuzzy(Word("hammer"), 3)),
))
self.assertEqual(str(tree),
'(foo:(bar[bazTOspam])ORNOT"ham ham"~2OR+hammer~3)')
self.assertEqual(
str(auto_head_tail(tree)),
'(foo:(bar [baz TO spam]) OR NOT "ham ham"~2 OR +hammer~3)',
)
# idem potent
self.assertEqual(
str(auto_head_tail(auto_head_tail(tree))),
'(foo:(bar [baz TO spam]) OR NOT "ham ham"~2 OR +hammer~3)',
)
def test_set_children(self):
test = self._test_set_children
test(Word("foo"), [])
test(Phrase('"foo"'), [])
test(Regex("/foo/"), [])
test(SearchField("foo", Word("bar")), [Word("baz")])
test(Group(Word("foo")), [Word("foo")])
test(FieldGroup(Word("foo")), [Word("foo")])
test(Range(Word("20"), Word("30")), [Word("40"), Word("50")])
test(Proximity(Word("foo")), [Word("foo")])
test(Fuzzy(Word("foo")), [Word("foo")])
test(Boost(Word("foo"), force=1), [Word("foo")])
many_terms = tuple(Word(f"foo_{i}") for i in range(5))
test(UnknownOperation(Word("foo"), Word("bar")),
(Word("foo"), Word("bar")))
test(UnknownOperation(*many_terms), many_terms)
test(AndOperation(Word("foo"), Word("bar")),
(Word("foo"), Word("bar")))
test(AndOperation(*many_terms), many_terms)
test(OrOperation(Word("foo"), Word("bar")), (Word("foo"), Word("bar")))
test(OrOperation(*many_terms), many_terms)
test(Plus(Word("foo")), [Word("foo")])
test(Not(Word("foo")), [Word("foo")])
test(Prohibit(Word("foo")), [Word("foo")])
def test_fuzzy_negative_degree(self):
check = LuceneCheck()
query = Fuzzy(Word("foo"), "-4.1")
self.assertFalse(check(query))
self.assertEqual(len(check.errors(query)), 1)
self.assertIn("invalid degree", check.errors(query)[0])
def test_combination(self):
tree = AndOperation(
OrOperation(
SearchField(
"mine",
FieldGroup(
Plus(AndOperation(
Word("foo"),
Regex("/fizz/"),
), ), ),
),
Boost(
Group(
AndOperation(
Phrase('"ham"'),
Word("spam"),
Prohibit(Fuzzy(Word("fuzz"))),
), ),
force=2,
),
),
Not(OrOperation(
Word('"bar"'),
Word('"baz"'),
), ),
)
to_path = simple_naming(tree)
paths_ok, paths_ko = self.propagate_matching(tree, set())
self.assertEqual(
paths_to_names(tree, paths_ok),
{"prohibit", "not"},
)
self.assertEqual(
paths_to_names(tree, paths_ko),
{
"and", "or", "searchfield", "fieldgroup", "plus", "and2",
"foo", "fizz", "boost", "group", "and3", "ham", "spam",
"fuzzy", "or2", "bar", "baz"
},
)
# adding matching just enough positive expressions, so that complete expression matches
paths_ok, paths_ko = self.propagate_matching(
tree,
{to_path["foo"], to_path["fizz"], to_path["ham"]},
)
self.assertEqual(
paths_to_names(tree, paths_ok),
{
"and", "or", "searchfield", "fieldgroup", "plus", "and2",
"foo", "fizz", "ham", "prohibit", "not"
},
)
self.assertEqual(
paths_to_names(tree, paths_ko),
{"boost", "group", "and3", "spam", "fuzzy", "or2", "bar", "baz"},
)
# making everything match
paths_ok, paths_ko = self.propagate_matching(
tree,
{to_path["foo"], to_path["fizz"], to_path["ham"], to_path["spam"]},
)
self.assertEqual(
paths_to_names(tree, paths_ok), {
"and", "or", "searchfield", "fieldgroup", "plus", "and2",
"foo", "fizz", "ham", "prohibit", "boost", "group", "and3",
"spam", "not"
})
self.assertEqual(paths_to_names(tree, paths_ko),
{"fuzzy", "or2", "bar", "baz"})
# making everything match, but some negative expression
paths_ok, paths_ko = self.propagate_matching(
tree,
{
to_path["foo"],
to_path["fizz"],
to_path["ham"],
to_path["spam"],
to_path["fuzzy"],
to_path["bar"],
},
)
self.assertEqual(
paths_to_names(tree, paths_ok),
{
"or",
"searchfield",
"fieldgroup",
"plus",
"and2",
"foo",
"fizz",
"ham",
"spam",
"fuzzy",
"or2",
"bar",
},
)
self.assertEqual(
paths_to_names(tree, paths_ko),
{
"and",
"boost",
"group",
"and3",
"prohibit",
"boost",
"group",
"and3",
"not",
"baz",
},
)
def test_fuzzy_non_word(self):
check = LuceneCheck()
query = Fuzzy(Phrase('"foo bar"'), "2")
self.assertFalse(check(query))
self.assertEqual(len(check.errors(query)), 1)
self.assertIn("single term", check.errors(query)[0])